By Abigail Mbah
Nigeria’s Independent National Electoral Commission (INEC) has removed artificial intelligence-generated photographs from its voter register, but the revelation that some of the questionable images were uploaded by the commission’s own registration officials has raised concerns about internal oversight and data integrity.
INEC officials confirmed that some personnel used their authorised credentials to upload manipulated photographs into the voter database. The commission has reportedly identified the individuals involved across about 10 states and commenced disciplinary proceedings against them, with some cases potentially facing prosecution.
The development emerged as part of a wider cleanup exercise that led to the removal of more than 1.6 million invalid records from the voter register. The affected records included duplicates, deceased persons and other irregular entries identified through the Automated Biometric Identification System (ABIS). AI-generated photographs were among the irregularities detected.
With the 2027 general elections approaching, the incident raises important questions about the integrity of the voter registration process, particularly how manipulated images passed through the commission’s verification procedures and why existing internal controls failed to prevent authorised personnel from introducing questionable records.
In response, INEC has introduced a new measure requiring voter photographs to be captured exclusively through an approved application, effectively ending the use of external webcams for the process. The move is intended to strengthen image verification and reduce opportunities for the submission of manipulated photographs.
The commission has also approved an Artificial Intelligence Governance Framework developed in collaboration with partners, including the Institute for Security Studies. The framework is expected to provide guidelines for the responsible deployment and management of AI technologies across INEC’s operations.
However, the effectiveness of these measures will depend largely on their implementation, particularly the ability to strengthen staff accountability, restrict unauthorised access and ensure that all voter registration activities are properly monitored.
INEC noted that the preliminary voter register is still subject to the ongoing claims and objections process, which provides an opportunity to identify and address irregularities before the register is finally certified.
Nevertheless, the involvement of some registration officials in uploading AI-generated photographs highlights a challenge that goes beyond technological vulnerabilities. It raises concerns about staff supervision, access management and the effectiveness of existing safeguards designed to protect the credibility of the electoral database.
The incident also demonstrates the growing risks artificial intelligence poses to public databases. As AI tools make it increasingly easy to generate realistic but fabricated images, organisations responsible for managing sensitive records must strengthen both their technological defences and internal accountability systems.
For INEC, the challenge is not just to remove fraudulent images but to ensure that the processes and personnel responsible for maintaining the voter register cannot be used to compromise its integrity ahead of the 2027 elections.



